A Unitree H1 humanoid robot malfunctioned during testing in China, leading to erratic movements and sparking fresh debate on AI safety. Experts attribute the incident to a programming issue, highlighting the ongoing need for rigorous safety protocols in robotics development.

In May 2025, Elon Musk’s forecast of a world with 10 billion humanoid robots continues to shape debate. As Tesla ramps up Optimus production and robotic surgery advances, the implications for society, industry, and AI regulation remain in focus.

Nvidia has announced a major deal to supply 18,000 Blackwell GB300 AI chips to Humain, a Saudi AI startup, for use in a new 500-megawatt data center. The agreement marks a significant step in Saudi Arabia’s Vision 2030, aiming to drive digital transformation and position the country as a regional leader in artificial intelligence.

Japan’s AIST has launched ABCI 3.0, a powerful supercomputer supporting cutting-edge AI research with 6.22 exaflops, advanced cooling systems, and over 75 petabytes of storage. Designed for multimodal AI, it facilitates collaboration with global researchers, pushing the boundaries of generative AI.

Houston is positioning itself as a major player in AI manufacturing, with Apple and Nvidia investing in local production facilities. These projects promise economic growth and job creation but come with challenges, including workforce retraining and energy infrastructure needs.

Automate 2025 returns to Detroit from May 12–15, spotlighting artificial intelligence in robotics across automotive, electronics, and logistics industries. Exhibitors including Universal Robots and MiR will showcase innovations in defect detection, precision assembly, and autonomous material handling.

Thinking Machines Lab, founded by ex-OpenAI CTO Mira Murati, is targeting a $2 billion seed round to develop adaptable, human-aligned AI systems. Despite lacking a public product, its $10 billion valuation reflects strong investor interest and positions the company as a notable entrant in the evolving AI sector.

Apple’s push into artificial intelligence, centered on Apple Intelligence and a revamped Siri, has faced delays stemming from GPU shortages, leadership reshuffles, and internal disputes. Budget constraints and missed deadlines have affected rollout plans, while a federal lawsuit and market downturn add to mounting pressure. The upcoming fall 2025 Siri release will be pivotal in restoring momentum.

Gemma 3 is Google’s latest suite of lightweight AI models, offering high performance on everyday devices. With built-in safety features and broad developer support, it empowers creators to build smarter, safer, and more accessible AI-driven applications for global use.

GXO Logistics is trialing AI-powered humanoid robots from Agility Robotics, Apptronik, and Reflex Robotics to enhance warehouse operations. These robots, including Agility’s Digit, are performing tasks like moving containers, helping reduce manual labor. While the technology is promising, challenges in flexibility and precision still limit widespread adoption.

Apple is accelerating its healthcare ambitions with Project Mulberry, an AI-powered virtual health coach integrated into its Health app. The initiative will deliver personalized advice using data from Apple devices and expert-backed guidance in sleep, nutrition, and fitness—potentially launching as the new Health+ service.

Extropic, a startup led by Guillaume Verdon and Trevor McCourt, is developing a novel silicon-based chip that uses thermodynamic fluctuations to perform probabilistic computations. Designed for energy-intensive tasks like Monte Carlo simulations, the chip claims to be 1,000 to 10,000 times more energy-efficient than current hardware. With ambitions to disrupt AI datacenter power consumption, Extropic's approach could mark a major shift in computing technology.

Cerebras Systems has launched the WSE-3, its latest AI-focused processor featuring 900,000 cores and built for training massive models. With 125 petaflops of performance and cutting-edge architecture, it powers the CS-3 supercomputer and targets industries from drug discovery to climate modeling.

On March 12, 2025, over 120 European tech founders announced Project Europe, a €10M fund supporting young entrepreneurs. Led by 20VC’s Harry Stebbings, the initiative offers €200K per startup, mentorship, and resources to drive AI and deep tech innovation while addressing Europe’s talent migration challenge.

On March 2, 2025, Firefly Aerospace’s Blue Ghost made history with the most precise AI-powered lunar landing by a private company. Using advanced vision-based navigation and hazard avoidance, the lander touched down near Mons Latreille, deploying 10 NASA payloads to advance lunar science. This milestone underscores AI’s growing role in space exploration, setting a benchmark for future missions to the Moon and beyond.

Giuseppe Di Morabito’s Milan Fashion Week debut on February 28, 2025, blended AI and couture with Ameca, the world’s most advanced humanoid robot, leading his Fall 2025 collection, Alone with the Stars. Inspired by a traumatic accident in Namibia, the show at Spazio Maiocchi fused personal reflection with medieval armor motifs and cutting-edge technology.

A new AI initiative in Victoria is bringing companionship to aged care homes through Abi, a humanoid robot designed to combat loneliness. Developed by Andromeda, Abi engages residents with multilingual conversations, emotion recognition, and playful interactions like dancing and bubble-blowing. As the trial progresses in Prahran, its success could shape the future of AI-driven social care while sparking debates on balancing technology and human connection.

Meta is making a bold move into AI wearables with six new devices set for 2025. As the company shifts focus from its struggling metaverse ambitions, it aims to revolutionize smart glasses and mixed reality. However, challenges like battery efficiency, market acceptance, and competition could impact its success. Will this be the future of AI-driven tech or another failed experiment?

A groundbreaking spintronic AI chip, developed by researchers from Tohoku University, integrates memory and computing functions, mimicking the human brain. This breakthrough in spintronics could lead to more energy-efficient AI hardware, reducing computational power demands while enhancing adaptability. However, challenges such as scalability and ethical concerns must be addressed for widespread adoption.

DeepSeek AI has emerged as a formidable challenger to OpenAI, boasting efficiency-driven breakthroughs in AI training while raising questions about its development costs, open-source credibility, and potential privacy risks. As experts debate the legitimacy of its rapid rise, reports suggest the company may have bypassed U.S. export controls for GPUs. This article explores the innovation, challenges, and controversies surrounding DeepSeek.